30 Single-Molecule Level Study and Control …
527
We studied the effect of localized plasmon on the FRET process by imaging
simultaneously dark-field scattering from the GNR and fluorescence from the donor
and acceptor molecules attached to the GNR in an optical microscope. From each
hybrid GNR, we measured scattering and fluorescence intensity as well as spectra
as a function of the polarization of incident light. Dark-field image of hybrid GNRs
is shown in Fig. 30.5c, with several bright spots corresponding to scattering from
individual particles. The same particles are identifiable in the corresponding fluorescence image in Fig. 30.5d. Orientation of the particles was determined from the
polarization dependence of scattering intensity for each GNR. Fluorescence imaging
of the GNRs was performed with 568 nm laser which efficiently excites the donor
molecules, but also overlaps with the longitudinal plasmon mode. When the excitation light is polarized along the long axis of the GNR, fluorescence spectra from both
the donor and acceptor are visible, as shown in Fig. 30.5e (blue dots). The fluorescence intensity of the acceptor is higher than that of the donor in this case. When the
excitation polarization is switched to perpendicular to the GNR long axis, the fluorescence intensities of both donor and acceptor drop, but the decrease is more prominent
for the acceptor, as seen in the spectra in Fig. 30.5e (red dots). Thus, switching of
the polarization of the excitation light by 90° changes the intensity ratio of the donor
and acceptor fluorescence, which is a signature feature of changing FRET efficiency.
To visualize the selective turn-on of FRET, we plotted the fluorescence spectra as
a function of the angle between the polarization of the excitation light and the long
axis of the GNR in Fig. 30.5f. Fluorescence intensity from the acceptor molecules
increases to a maximum when the polarization is parallel to the long axis of GNR
and varies periodically with the polarization angle. Ratio between the fluorescence
intensity of the acceptor and the donor (shown in Fig. 30.5g) changes periodically
between 2 and 0.5 for the two orthogonal polarizations.
To quantitatively evaluate the results of the fluorescence spectral changes in terms
of FRET efficiency due to the GNR localized plasmon, we used a theoretical model
which accounts for the enhancement of FRET as well as for quenching of the molecular excitations by the GNR. We approximated the hybrid GNR with a single donor
dipole moment and a single acceptor dipole moment. The calculations were carried
as fitting of the experimental fluorescence spectra. Results of the fitting are shown
by solid lines in Fig. 30.5e. The calculated efficiency of the energy transfer in the
presence of the GNR was E = 0.46 with a Förster radius of R 0 = 15.04 nm. To evaluate the effect of the plasmon enhancement, we kept the geometry of the donor and
acceptor dyes and carried out the calculations without the GNR, with resulting FRET
efficiency of E = 0.01 and Förster radius of R 0 = 7.17 nm. From this comparison, we
obtained FRET enhancement factor due to the presence of GNR of 46. An average
FRET efficiency enhancement obtained from 10 nanoparticles was by a factor of 65.
At the same time, the Förster radius on average increased twofold in the presence of
the GNR.
527
We studied the effect of localized plasmon on the FRET process by imaging
simultaneously dark-field scattering from the GNR and fluorescence from the donor
and acceptor molecules attached to the GNR in an optical microscope. From each
hybrid GNR, we measured scattering and fluorescence intensity as well as spectra
as a function of the polarization of incident light. Dark-field image of hybrid GNRs
is shown in Fig. 30.5c, with several bright spots corresponding to scattering from
individual particles. The same particles are identifiable in the corresponding fluorescence image in Fig. 30.5d. Orientation of the particles was determined from the
polarization dependence of scattering intensity for each GNR. Fluorescence imaging
of the GNRs was performed with 568 nm laser which efficiently excites the donor
molecules, but also overlaps with the longitudinal plasmon mode. When the excitation light is polarized along the long axis of the GNR, fluorescence spectra from both
the donor and acceptor are visible, as shown in Fig. 30.5e (blue dots). The fluorescence intensity of the acceptor is higher than that of the donor in this case. When the
excitation polarization is switched to perpendicular to the GNR long axis, the fluorescence intensities of both donor and acceptor drop, but the decrease is more prominent
for the acceptor, as seen in the spectra in Fig. 30.5e (red dots). Thus, switching of
the polarization of the excitation light by 90° changes the intensity ratio of the donor
and acceptor fluorescence, which is a signature feature of changing FRET efficiency.
To visualize the selective turn-on of FRET, we plotted the fluorescence spectra as
a function of the angle between the polarization of the excitation light and the long
axis of the GNR in Fig. 30.5f. Fluorescence intensity from the acceptor molecules
increases to a maximum when the polarization is parallel to the long axis of GNR
and varies periodically with the polarization angle. Ratio between the fluorescence
intensity of the acceptor and the donor (shown in Fig. 30.5g) changes periodically
between 2 and 0.5 for the two orthogonal polarizations.
To quantitatively evaluate the results of the fluorescence spectral changes in terms
of FRET efficiency due to the GNR localized plasmon, we used a theoretical model
which accounts for the enhancement of FRET as well as for quenching of the molecular excitations by the GNR. We approximated the hybrid GNR with a single donor
dipole moment and a single acceptor dipole moment. The calculations were carried
as fitting of the experimental fluorescence spectra. Results of the fitting are shown
by solid lines in Fig. 30.5e. The calculated efficiency of the energy transfer in the
presence of the GNR was E = 0.46 with a Förster radius of R 0 = 15.04 nm. To evaluate the effect of the plasmon enhancement, we kept the geometry of the donor and
acceptor dyes and carried out the calculations without the GNR, with resulting FRET
efficiency of E = 0.01 and Förster radius of R 0 = 7.17 nm. From this comparison, we
obtained FRET enhancement factor due to the presence of GNR of 46. An average
FRET efficiency enhancement obtained from 10 nanoparticles was by a factor of 65.
At the same time, the Förster radius on average increased twofold in the presence of
the GNR.